Pretraining Language Models via Neural Cellular Automata

(hanseungwook.github.io)

30 points | by shmublu 3 days ago

4 comments

  • benob 8 minutes ago
    Reminds me of "Universal pre-training by iterated random computation" https://arxiv.org/pdf/2506.20057, with bit less formal approach.

    I wonder if there is a closed-form solution for those kinds of initialization methods (call them pre-training if you wish). A solution that would allow attention heads to detect a variety of diverse patterns, yet more structured than random init.

  • dzink 1 hour ago
    “The long-term vision is: foundation models that acquire reasoning from fully synthetic data, then learn semantics from a small, curated corpus of natural language. This would help us build models that reason without inheriting human biases from inception.”
    • qsera 45 minutes ago
      I think this is a bit risky, because it assumes that all knowledge that a human posses about nature is acquired after birth.

      But is that correct? I think organisms also come with a partial built in understanding of nature at birth.

      • throw-qqqqq 8 minutes ago
        > I think organisms also come with a partial built in understanding of nature at birth

        I agree. Most organisms are quite pre-trained: they have “instincts” and natural behaviors.

        E.g. newly hatched turtles know to crawl towards the ocean immediately when they hatch. They don’t learn that on their way.

        It seems to me that most lifeforms come into this world pre-trained.

      • jamilton 11 minutes ago
        I don’t think that assumption is being made, why do you think that? In terms of metaphor, training a model could be considered both knowledge acquired after birth and its evolution. But I don’t think it’s particularly useful to stay thinking in metaphors.
  • Heer_J 9 minutes ago
    [dead]
  • voxleone 1 hour ago
    Neural cellular automata are interesting because they shift learning from “predict tokens” to “model state evolution.” That feels much closer to a transition-based view of systems, where structure emerges from repeated local updates (transitions) rather than being encoded explicitly.

    I'm working on a theoretical/computational framework, the Functional Universe, intended for modeling physical reality as functional state evolution. i would say it could be used to replicate your CA process. Won't link it here to signal my good faith discussing this issue - it's on my GH.